The proposed work seeks to understand better the properties of the membrane ion channels activated by glutamate, the main excitatory transmitter in the mammalian central nervous system (CNS). The studies will focus on glutamate-receptor (GluR) channels in granule neurons from the rat cerebellum. Two main approaches will be used, both of which will employ patch-clamp techniques. One approach will be to study the native channels in primary explant cultures or thin slices of rat cerebellum. The other approach will involve similar experiments on the channels encoded by GluR cDNAs after expression of these channels in mammalian cells or Xenopus oocytes. One primary goal of the studies on native channels will be to extend the characterization of the fS-conductance channels in granule cells. These channels are: selectively activated by glutamate and kainate; appear to be the main type of GluR channel in granule cells; and can be studied in relative isolation in cultures from young animals. Experiments will be done to determine the gating kinetics, ion selectivity, and pharmacology of these channels. Other experiments will focus on native GluR channels that are selectively activated by N-methyl-D-aspartate (NMDA) and will test the hypothesis that 'clustering' of NMDA-receptor (NMDAR) channel activity is influenced by phosphorylation of the channel. The studies using cDNA clones will be limited to GluR subunits thought to be expressed in cerebellar granule neurons, as indicated by the expression of the corresponding mRNA. Whether GluR mRNAs known to be present in the granule cell layer of the cerebellum are indeed expressed in granule cells will be determined by performing Northern blot analysis of RNA extracted from purified populations of granule cells; whether the cells express the edited or unedited version of GluRs 5 and 6 (Bettler et al., 1990; Egebjerg et al., 1991) will be tested by using strategies that employ version-specific oligonucleotides (Sommer et al., 1991). The GluR subunits so localized in granule cells will then be expressed for study in patch-clamp experiments to determine the conductance and gating kinetics of the channels. Preliminary reports from other laboratories indicate that the conductance of the channels formed from non-NMDA subunits is too small (< 1 pS) to measure single-channel currents directly; and estimates of their properties will therefore be obtained from analysis of agonist-evoked current noise, both in cells and in excised membrane patches. Other experiments will seek to measure single- channel currents through heteromeric channels formed by NMDAR subunits. This work will further elucidate the properties of GluR channels and the functional consequences of their subunit composition. Because of the importance of glutamate in the mammalian CNS, both in fast synaptic transmission and in mediating long-term changes in synaptic strength, such information is fundamental to an understanding of brain function.